A rational design philosophy for long distance slurry pipelines

Page 1

Reprint

of Paper published in

Li

CHEMICAL ENGINEERING IN AUSTRALIA

THE INSTITUTION OF ENGINEERS, AUSTRALIA

L

Page 2

22 LONG DISTANCE SLURRY PIPELINES - Thomas

A Rational Design Philosophy for Long

Distance Slurry Pipelines.

A.D. THOMAS

Various types of slurries ranging from ones with very coarse particles to very fine particles are fine particles. The differences between typical tailings line slurries in mineral processing plants and the It is shown how pressure drop limitations dictate that all long distance slurries have quite true long distance slurry is discussed and it is shown how the stability of a slurry on shutdown of the slurry. It is argued that the stability of a slurry depends on the yield stress of the slurry being pipeline and its ability to be restarted is perhaps the most important requirement for a long distance variables. This requirement and the limitations on maximum economic pressure drop and velocity dictate the sufficiently high to support the largest particles and a simple theory is developed relating these two

available size distribution limits for long distance slurries. Finally economic optimisation of the operat-

ing velocity, solids concentration and solids throughput is discussed with the aid of a typical example.

Tex extrapolated yield stress

The hydraulic transport of solids in pipelines is Tst static yield stress not new having been used for short distance trans- 3 HYDRAULIC BEHAVIOUR OF SLURRIES port in the mining industry for over a hundred years. In the last decade a few long distance (up The behaviour of different types of slurries in to 440 km) pipelines have been built (see Table 1) horizontal pipe flow is best illustrated by conand there is increasing interest being shown in sidering a graph of pressure gradient versus mean this alternative transport means. pipe velocity as in Figure 1 which shows the

behaviour of slurries having the same concentration

Because long distance pipelines are relatively new but different particle size in two pipe sizes. Four the published information relating to the hydraulic types can be distinguished (see Thomas (1976) design criteria and methods is scarce and generally according to the different hydraulic behaviour. vague about details. The majority of research effort and publications have tended to be concerned 3.1 Type 1 Slurry with the types of slurries which have traditionally been pumped ie., those pumped in the short distance This is a typical "settling" slurry having relative applications and this information is of little help ly coarse particles. Examples are gravel or coarse in the design of a long distance pipeline. sand. These are often described as heterogeneous

slurries because of the distinct vertical concen-

It will be shown here how certain unique requiretration gradient which exists ments of long distance slurry pipelines dictate the the deposit conditions. At sufficiently high vel-

same requirements also ocities they will flow as pseudo-homogeneous

dictate the solids concentration and particle sizes slurries as shown by the straight line portion of which can be pumped and the pumping velocity. A the graph. The deposit velocity, defined here as discussion of these three variables shows how the the velocity below which a stationary bed of solids available range for them is limited. This explains begins to accumulate on the bottom of the pipe, is for a particular commodity, the solids concencrucial to design as it is generally desirable tration, particle size and pumping velocity of long operate above this velocity. For these types of distance slurries does not vary much from installaslurries Durand (1953) has shown that the deposit tion to installation. velocity varies as the square root of pipe diameter.

This means that the operating velocity becomes

2 NOTATION very large for large diameter pipes with consequent

high pressure gradients. The pressure gradient of

solids concentration by volume unless otherslurries can be estimated using the Durand wise stated equation (Durand (1953)). internal pipe diameter particle diameter or mesh size 3.2 Type 2 Slurry gravitational acceleration, 9.81 As the particle size is reduced the pseudo-homogenconstant in equation A2 eous portion of the curve extends to lower velocit- AP/L pressure gradient ing" or "semi-heterogeneous". For sand, ies and the slurry can be termed "moderately typical settlvolume flow rate mean velocity in pipe median particle sizes might lie between 100 and 500 pl plastic viscosity microns. Thomas (1976) has shown how the pressure density of solids-liquid mixture gradient of these slurries cannot be scaled up by density of solid particles use of the Durand equation and has suggested an Dr. Thomas is Senior Research Engineer, M.D. alternative approach. It has further been shown (Thomas (1975)) that for these slurries Research Company, North Ryde, NSW (Paper H1001 submitted 10 December 1976)

The Institution of Engineers, Australia

Page 3

LONG DISTANCE SLURRY PIPELINES - Thomas 23

TABLE 1.

SOME LONG DISTANCE SLURRY LINES

Location Material S.G. Length Pipe Max. *Particle Reference

(km) Diameter Particle Size (mm) Size (mm) Limit (mm)

Savage River Magnetite 5.0 85 225 0.104 0.30 McDermott et al (1969) 2 Bougainville Copper 4.3 27 150 0.208 0.40 Piercy (1976) Concentrates 3 West Irian Copper 4.3 110 100 0.149 0.40 McNamara (1976) Concentrates 4 Arizona (Black Coal 1.4 437 450 2.38 3.0 Wasp (1969) Mesa) 5 Ohio Coal 1.4 173 250 1.41 3.0 Halvorsen (1976) 6 England Limestone 2.7 112 250 0.59 1.75 Schriek et al (1973/1) 7 Japan Copper 2.7 65 200 0.208 0.70 Couratin (1969) Tailings 8 Turkey Copper 4.3 64 125 0.149 0.40 Noyan et al (1974) Concentrates 9 Washington Limestone 2.7 67 200 0.295 1.75 Schriek et al (1973/1) 10 Brazil Iron ore 5.0 403 500 0.30 Anonymous (1975) *Calculated as per Section 6. Note numbers 6 and 9 flow in laminar flow regime Number 10 is not yet operating 1.0 equation will overpredict. For pipe sizes in the

range 25 to 100 mm, where most the research work

Solids S.G. 2.65 is performed, the Durand equation can adequately

• 50 mm pipe slurries is less than for the coarser predict the pressure gradient. "settling"

The deposit velocity for these "moderately settling"

So slurries but is still proportional to Do.5. Thus high operating velocities are needed in large pipes. Typical examples of such slurries are sand mining The iron sand slurry ship loading schemes at Waipipi (Thompson et al 1972)) and Taharoa (Raudkivi (1975)) in New Zealand are further

(m water/m) • 10.05 pipe wear rates. and dredging applications and some coarser tailings. examples. In these latter two cases velocities rethe consequent high pressure gradients and high quired in the 300mm pipes are around 5m/s with Pressure Gradient • 005.01 2 pseudo-homogeneous behaviour is maintained right is not only lower than for slurries 1 and 2 but is As the particle size is further reduced eventually down to the deposit velocity as illustrated in also less dependent on pipe size. Probably the best Figure 1. For these slurries the deposit velocity from which it can be shown is due to D.G. Thomas (1961). He gives an equation available for predicting the deposit velocity

Water velocity is roughly proportional to Do.1. This work was based on experiments on flocculated thorium oxide and kaolin slurries in pipes from 25.001 to 100 mm. Experimental data on iron ore slurries • 2 • 5 10 (Schriek et al (1973/2) and coal slurries (Schriek

et al (1973/3)) in pipe sizes from 50 to 300 mm

Velocity (ms -1, shows a similar dependence of deposit velocity

Figure 1. Typical behaviour of different pipe diameter, as does unpublished data obtained by

a number of different

types of slurries pipe sizes from 19mm to 105 mm. This means that the accuracy of the prediction using the Durand velocity and hence the operating velocity can be even for very large diameter pipes the deposit equation depends on the pipe size under considerfairly low. ation. For small pipe sizes (less than about 25mm) the Durand equation will underpredict while for Because deposition occurs with this type of slurry large pipe sizes (above about 100 mm) the Durand under turbulent flow conditions it means that if

Chemical Engineering in Australia, 1977

Page 4

24 LONG DISTANCE SLURRY PIPELINES - Thomas

any non-Newtonian effects are present they are relatively insignificant. This means that for suitable long distance slurries are often the design purposes the pressure gradient may be estipseudo-stable variety. mated using a Newtonian fluid approach using either 4 EFFECT OF PARTICLE SIZE DISTRIBUTION ON calculated viscosity along the lines suggested by the measured viscosity of the slurry or else the HYDRAULIC BEHAVIOUR approach which is extensively used in industry is Vocadlo and Charles (1972). An even simpler When considering the different types of slurries it by the ratio of the density of the slurry to that to simply multiply the water only pressure gradient should be noted that not only is the median 50) particle size important in determining slurry behaviour but also the particle size distribution. basis of the of water. This latter approach is of course the slurry pump engineer's measurement of For essentially mono-sized particles of sand of pressure drop in "head of slurry". S.G. 2.65 typical ranges of particle size might be as follows: Typical examples of type 3 slurries are the usual tailings slurries pumped anywhere from a few Type 1 d > 0.50 mm hundred metres to 10 or 15 km. Pumping velocities Type 2 Type 3 0.010 < d < 0.10 < d < 0.10 mm 0.50 mm for such slurries are generally around 1 to 2m/s. Type 4 d < 0.010 mm manner at all velocities above the deposit velocity Although these slurries flow in a homogeneous With the more usually encountered slurries having they are termed pseudo-homogeneous because if the wide size distributions with particles perhaps sub-micron to mm or em size no typical d from coarsest particles reaching the bottom first and flow is stopped they will quickly settle out, the can be given since the degree of homogenéity, or the finest last, to form a bed of solids on the proportion of micron size particles, the maximum conversly heterogeneity, will depend on the

particle size, the concentration and the degree of

It can perhaps be noted here that the pressure wide size distribution could exhibit all four types flocculation. For example, a flocculated slurry of gradient prediction method presented by Wasp et al is an attempt to provide a single equation of behaviour depending on the concentration, for types 1, 2 and 3 slurries. Their method thereto a type 4 slurry at high concentrations. changing from a type 1 slurry at low concentrations as the particle size is reduced and the slurry fore allows for a smooth reduction in pressure drop a certain Alternatively a slurry of wide size distribution at concentration could change from a type 1 changes from a type 1 to a type 3. to a type 4 depending on the degree of 3.4 Type 4 Slurry flocculation. As the particle size is further reduced to the The method developed by Wasp et al (1970) which classifies slurries according to the distribution particles are colloidal in size the behaviour of stage where a significant proportion of the of solids concentration within the pipe may be of With these the slurry if flow is suddenly stopped changes. of a slurry. some use in quantifying the degree of homogenity slurries no settling of particles occurs, or if it does, the particles settle (or compact) en masse forming a homogeneous but more concentrated 5 DESIGN CRITERIA FOR LONG DISTANCE SLURRIES slurry with clear water above. The completely nonsettling type of slurry will be referred to as a Because of the high velocities required and the fully stable slurry while the latter type, where type 1 and 2 slurries, pumps are required at more consequent high pressure gradients when pumping will be referred to as a pseudo-stable slurry. settling occurs but with no particle segregation, frequent intervals along the pipeline. overall capital and operating costs then make it The higher if flowing under laminar conditions, Both are defined as type 4 slurries. uneconomic to pump these slurries over long described by non-Newtonian rheological models which pumped at about 5 ms For example the Waipipi iron sands in a 300 mm pipe has a Aziz (1972). They will generally possess a yield are discussed in numerous textbooks e.g. Govier and (Thompson et al (1972)) which is about three times pressure gradient around.075 m water/metre stress and can often be modelled as the pressure gradient experienced in the smaller slurry would be as shown in Figure 1. For a pseudo- Typical behaviour of a fully stable et al (1969)). (230 mm) Savage River iron ore pipeline (McDermott For long distances we are therefore especially in larger pipe sizes. Thus the fact that stable slurry laminar flow is not always possible, limited to type 3 or type 4 slurries. discussion has shown how both these The previous stopped does not necessarily mean that laminar flow a slurry is pseudo-stable, flow is suddenly slurries have similar pressure gradients in the turbulent flow regime so that it would later. is possible. This matter will be further discussed first sight that either type would be suitable for

long distance applications. Indeed most tailings

Under turbulent flow conditions the behaviour of long distance pipelines. lines operate at similar pressure gradients to the However consideration of type 4 slurries will closely parallel the type 3 the following requirements of a long distance slurries but will generally exhibit a slightly higher pressure drop because of their more viscous slurry will indicate that in fact the only slurry nature. suitable for long distance pipelines is the type 4 In addition it will be shown how even with- There is a vast amount of literature published type 4 category there is only a limited slurries which are economically suitable. concerning pressure drop prediction of non-Newtonian Govier and Aziz (1972) and a recent review by Cheng homogeneous slurries, see for example the book by The major design requirements for a long distance slurry are: tween fully stable and pseudo-stable slurries but This literature does not distinguish be- (a) It must be capable of being stopped for at is in fact only applicable to the former. Unleast a few days and restarted with full of slurry. the pipeline fortunately, as will be recognized later, the most (b) Even the coarsest particles must not deposit

The Institution of Engineers, Australia

Page 5

LONG DISTANCE SLURRY PIPELINES - Thomas 25

out at the normal operating velocity for fear of a pipelines should ideally employ type 4 slurries (c) The velocity should not exceed about 1.75 ms l slow build up causing a blockage. which do not settle out to form a bed of solids which have been previously described as those since, as noted by Wasp (1969), this velocity has when stopped. been found to be about the economic maximum with As discussed previously a type 4 slurry can be single phase fluids such as with oil pipelines. either fully stable or pseudo-stable. Obviously causes capital and running costs to rapidly become Above this velocity the high pressure gradients a fully stable slurry which remains completely In any case at velocities much above homogeneous even after days of standing will not present any start up problems. A pseudo-stable this wear due to abrasion could become increasingly slurry can also be restarted as has been inferred

by Wasp (1969) when he refers to fast or slow

These three requirements will now be discussed in settling slurries being permissible as long as the detail. material settles homogeneously.

To determine whether a slurry will be stable (fully

5.1 Restart Capability pseudo) a simple stability criterion has been

That the slurry must be capable of being restarted assumption that the coarsest particles will not developed (see Appendix).

after stopping is obvious even though such situations may never be envisaged in normal slurry yield stress exceeds the gravitational settle providing the upwards force due to the It is often general practice to fill For a particular yield stress (either the line with water before any planned long periods static or extrapolated see Appendix) this criterion of shutdown but the possibility must be allowed for enables the maximum stable particle size to be that an unplanned shutdown could occur. calculated. Conversely for a given maximum part- 3 slurry is stopped flowing the solids will quickly settle out, the coarsest particles reaching the icle size it permits calculation of the required first and the finest last. Thus a bed of yield stress. The yield stress of a slurry solids will be formed filling perhaps 40 or 50% of largely due to the interaction between colloid or the pipe cross section. In a horizontal pipe this near colloid size particles. bed should present no problem since once flow means that the slurry must have a wide size

A typical distribution for a coal

started the water flowing to 10% greater than 1mm ranging progressively scour the bed away. situation is different. down to 15 to 30% less than 50 microns. The possibility exists that a bed of particles may The importance of the yield stress in determining cross section at a the slope and block up the whole pipe "valley" section. slurry stability has been clearly illustrated by possibility is that the same blockage could arise tests by the author (Thomas (1975)) on a finely ground quartz slurry. This was a ball mill proion flow effect as discussed by Shook (1974). from density currents which are a natural convectduct which, as produced, was largely dispersed such a blockage does occur starting of the pipeline (de-flocculated) and possessed virtually no yield If allowed to settle this slurry quickly will require that this plug be moved. stress. formed a solid bed of particles which were packed Televantos (1976) have shown that it is possible to so hard that the bed could oftenonly beremoved by start such plugs and indeed pump continuously at a considerable amount of chipping with a hammer required is very high. such high concentrations but the pressure gradient For closely graded sand and chisel. However when the pH of the slurry they measured and predicted pressure gradients of was lowered causing flocculation the slurry developed a yield stress and became a completely the order of 1 m water/m regardless of pipe size stable slurry which did not settle and remained which is some 50 to 100 times higher than the typical tailings line pressure gradient. The fluid for weeks. situation with wide size distributions is likely The yield stress which type 4 slurries possess to be worse since they tend to pack tighter than must be overcome to restart a stopped pipeline. narrow size distributions. Clearly it would not For a completely homogeneous type 4 slurry which require a very long length of plugged line before does not settle at all on shutdown it is the yield the start up pressure required was beyond the stress of the slurry as pumped which must be overcapability of the pumps. come to restart the pipeline. pseudo stable In spite of this it is common knowledge that many type 4 slurry which compacts on shutdown there is the possibility that this compacted slurry will form a tailings lines, the solids of which settle out to bed on shutdown, can usually easily be rein the low "valley" sections of the pipeline started. However every now and again these taileither by sliding or by density currents.

restart the pipeline it is necessary to overcome

ings lines do block up but because of their short the yield stress of this compacted slurry in the lengths and because they are usually situated plugged sections. This compacted slurry, because above the ground it is not too difficult to of its higher solids concentration will have a locate the blockage and clear the line. words it is far better to accept the possibility of higher yield stress than the slurry as pumped. the line blocking up occasionally than to go to the In horizontal lengths of pipe the clear water additional cost of grinding the tailings finer just layer on top would be expected to begin moving to achieve a type 4 slurry for transportation. first since it requires no yield stress to (The very concept of tailings disposal requires initiate motion. As it moves it would pick up the that the solids settle out at the disposal compacted portion and dilute it and so eventually

as long as the yield stress of the pumped slurry was overcome the line would start moving again.

With a long distance slurry pipeline perhaps some hundreds of kilometres in length and with the The possibility exists, however remote, that upon pipeline buried under the ground even the remotshutdown, all of the compacted slurry moves to the est possibility that a blockage can occur must be lower "valley" sections of the pipeline so that avoided. For this reason all long distance slurry the whole pipeline consists of plugs of compacted

Chemical Engineering in Australia, 1977

Page 6

26 LONG DISTANCE SLURRY PIPELINES - Thomas

slurry separated by plugs of clear water. sections require no pressure to but the compacted slurry sections require that sufficient pressure be applied to everywhere exceed the yield stress in these sect- As an example suppose a coal slurry is pumped at 40% volume concentration and that this slurry has an extrapolated yield stress of 3 Pa. Suppose

to a homogeneous slurry occupying only 80% of the shutdown this slurry slowly compacts down AP/I

volume ie., there is a clear water layer Log

occupying 20% of the volume. the yield stress of this 50% concentration compacted slurry is 8 Pa. The worst possible situation occurs if 80% of the line length consists of plugs of compacted slurry requiring 8 Pa stress to start UNSTABLE flow and 20% of the line is clear water requiring no stress to start flow. sections the shear stress at of the pipe on commencement of flow, Log Velocity

4L Figure 2. Effects of stability considerations

on available operating region.

so that the total pressure required before flow will commence is given by 5.2 No Deposition of Particles During Operation AP = 0.8 32 L This is the second design requirement. stability criteria given by equations A6 and A7 in The D the Appendix should ensure either a stable or

pseudo-stable slurry on sudden shutdown. However,

Providing this pump pressure is available the pipewhereas a fully stable slurry can be operated in

line should be able to restart under any conditions. the laminar flow regime, a pseudo-stable slurry

This would suggest that the grade limitations often cannot. This can be explained by the breakplaced on the earlier lines (for instance Savage down of the floc structure under conditions of River was limited to a maximum of 10% grade laminar shear. It is generally agreed (see McDermot et al (1969)) may be unnecessarily Michaels & Bolger (1962/2)) that the floc structure conservative. Possibly this was later realised is broken during laminar shear, indeed this explains because it was stated by E. Wasp (Wasp, (1969)) that "We put a slope limitation of somewhere bethe reduction in apparent viscosity with shear rate of a Bingham plastic. The broken increasing floc tween 10 and 15%, but not always. We are building structure is less able to support the coarsest one line now without a slope limitation". Further particles thus causing deposition. This is more evidence is provided by McNamara (1976) concerning likely to happen in large pipe sizes because, in the copper concentrate pipeline in Irian Jaya. He small pipe sizes, an apparent bridging effect grade of 18% but that during construction states that this line was designed to a maximum several of the floc structure. Figure 3 provides evidence (Michaels & Bolger (1962/1)) increases the strength grades of 26% and 28% were built. No start-up problems are mentioned. the more stable the slurry the less like-

further consideration is that with a pseudo-stable steep grades going to be of importance. 3

slurry which compacts on shutdown the degree of compaction may be influenced by the weight of 2 sections the concentration of the compacted slurry at the bottom could be greater than when tested in a bench experiment. This would need to be investivelocity The laminar flow curves of the type 4 slurry in gated for each particular slurry. slurry above so that in long steeply sloping pressure gradient to become independent of Figure 1 illustrate the tendency for the laminar only on the yield stress and independent of the in large pipes ie., to become dependent Velocity (ms"') limits on observed deposition Error bars indicate upper and lower aminar -transition - plastic viscosity of the Bingham model. Thus for • 3 laminar flow curves can be drawn as horizontal lines illustrative purposes in large pipe sizes the 10 50 100 200 of constant pressure gradient. Pipe Diameter (mm) The requirement that the slurry be stable on shut- Figure 3. Variation of deposit velocity down means that a certain minimum yield stress is with pipe diameter for a fast required which depends on the maximum particle size. compacting slurry. Thus the dashed lines on Figure 2 show permissible idealised laminar flow curves to ensure a stable of this phenomenon. It shows experimental results slurry. Consideration of the second and third obtained by the author for a loam slurry in two design requirements means that an upper limit is pipe sizes, 19 and 105mm. The loam consisted of placed on the yield stress. about 20% clay with the remaining sand having a top size of 0.82 mm. In the smaller pipe laminar/

The Institution of Engineers, Australia

Page 7

LONG DISTANCE SLURRY PIPELINES - Thomas 27

turbulent transition was observed at 1.4 ms At stress, T anging tabout 3 Pa for a coal slurry, of about 4.5 pa ar iron or slurry Both (visually through transparent viewing section). steady Laminar flow without deposition was these depend slightly on the pipe diameter. Note observed. At 0.8 ms a bed of solids formed. Thus that the turbulent pressure gradient is not greatly 0.8 and 1.3 ms the critical deposit velocity lay somewhere between › well below the laminar/turbulent dependent on differ for different solids S.G. 's. the yield stress although it will As an example, transition velocity. The same slurry, at the same Piercy (1976) has stated that the yield stress for concentration, in the larger pipe behaved differentthe Bougainville copper concentrate slurry (solids ly. In this case deposition occurred under S.G. 4.3) is maintained below a maximum limit of turbulent flow conditions. By scaling up from the 4.4 Pa which is in agreement with the above stated laminar flow results obtained in the smaller pipe limits. The lower limit for stability depends on (see Bowen (1961/1) it was estimated that the the maximum particle size. 1aminar/turbulent transition velocity in the large pipe was about 1.0 ms although laminar flow If the slurry is completely stable, operation withconditions could not be realised. in the laminar regime is feasible. For example the

Rugby limestone pipeline in England operates under

The slurry of Figure 3 illustrates typical pseudolaminar conditions. (Shriek et al (1973/1) In stable slurry behaviour which has been observed by this case the same economic limits on maximum the author with a number of different slurries. suddenly stopped the coarsest pressure gradient apply so that the upper limit can be raised as shown in Figure 4. This limit particles remain Sustained laminar would be about 2.5 times the previous limits ie., possible in small pipe sizes and in coal. = 11 Pa for iron ore down to 7.5 Pa for large pipes deposition occurs under conditions of at velocities above the laminar/ turbulent transition velocity. For pipe sizes Operating point for both large enough for deposition to occur under laminar and turbulent flow

conditions of will vary as pourbulent as for a type 3 slurry. flow the deposit velocity

The author knows of no method of predicting a Upper limit for economic - priori whether deposition will occur under laminar laminar operation or turbulent flow conditions. For rapidly settling pseudo-stable slurries it would therefore seem necessary to perform pipe loop tests in increasingturbulent regime. Once this size pipe is reached obtained by assuming it varies as Do.1 ly larger pipe sizes until deposition occurs in the deposit velocity in larger pipe sizes can as for a be the D AP/4L Upper limit for economic turbulent operation

type 3 slurry. Log

5.3 The Velocity Should Not Exceed About 1.75 ms l

Lower limit for

Operation above 1.75 ms--is generally uneconomic stability due to the rapid increase in pressure. an upper limit on the yield stress. uncertainty associated with whether or not laminar 0.5 N flow is possible it is generally desireable that the pipeline be operated in the turbulent regime. Velocity (ms) To ensure that operation is fully within the Figure 4. Yield stress limits for turbulent regime and to allow for possible variatlaminar and turbulent flow

prediction methods the transition velocity should ion in slurry properties and inaccuracies in operation.

a safety margin of 0.75 ms Note that this, then allows 6. CONSEQUENT LIMITS ON MAXIMUM PARTICLE SIZE turbulent regime as discussed above. of deposition occurring in the For example Considering only turbulently flowing slurries, for the deposition velocity for the slurry of Figure 3 of they are by far the most common, the upper limits = 4.5 Pa for iron ore and 3 for coal mean of 1.5 to 1.75 ms approaches thl for pipe diameters greater than usual operating velocity range stable slurry is (from equation A7 of the Appendix) that fhe maximum possible particle size for a about 1000 mm so for pipes smaller than this there about 0.3 mm for iron ore ranging up to about 3 mm would be no problem if design were based on the for coal. Of course a slurry can operate with laminar transition velocity. In short, as long as maximum particle sizes considerably less than the slurry does not compact very fast on shutdown these limits and may have to to prevent deposition based on the laminar transition velocity. (say within a few minutes), the design can be during operation but there is a limit to how fine fast compacting pseudo stable the particles can be ground for pipelining due to large diameter pipes it would be wise to perform pressure gradient as size is reduced due to increasthe rapidly increasing yield stress and hence small pipe loop tests to check whether deposition does occur in the turbulent regime, and then scaleing colloidal effects. For example the work of

B. by assuming the deposit velocity varies as Friend and Hunter (1971) suggests that the yield

stress increases as the square of the reciprocal of the particle size for colloidal size particles.

The allowable limits on T are now Table 1 gives some details of nine existing long Figure 4. Note that by plitting DAP/4L versus V distance slurry pipelines together with the maxi- Figure 4 is applicable to all large pipe sizes. maximum limit on The mum economic particle size as calculated above. In upper limit dictated by the 1 ms every case the maximum particle size is less than transition velocity represents a maximum yield

Chemical Engineering in Australia, 1977

Page 8

28 LONG DISTANCE SLURRY PIPELINES - Thomas

the calculated maximum limit the actual values ranging from one fifth to four fifths of the limit. of Elliott & Glidden's paper will reveal that the reasonable for long distance applications. long distance slurries would appear to lie between The economic limits on maximum particle size for with the 12.5 mm top size was around 10 Pa. (at extrapolated yield stress ensuring a stable slurry using T 20% and 100% of the value given by equation A7 equal to 4.5 Pa for iron ore ranging 59% concentration by weight). down to to Pa for coal. equation A7. agreement with the 11 Pa limit indicated by Note, however that it is higher than LIMITS ON CONCENTRATION economic limit and so may prove unecon-

omic for long distances. Similarly for a 3.2 mm

Most reports on slurry pipeline installations tration was stable ( T size they found that a 57.2% by weight concenspecify upper and lower limits to the permissible range of concentration. For example Halvorsen latter is difficult to ascertain but centration was not stafle. The yield stress of the line as being 50% and 44% by weight. (1976) mentions the limits on the Ohio coal pipethat the yield stress approximately doubled for these limits should now be obvious and following This would suggest a yield stress of about 2.7 Pa. with every 5% increase in concentration by weight. calculate the allowable limits for any particular the reasoning outlined above it is possible to It is remarkable that equation A7 indicates a

value of 2.9 required for stability.

The upper limit is set by the necessity to stay Thus it can be seen that slurries of virtually any top size can be made stable providing the yield coal, 4.5 Pa for iron ore). below the maximum economic yield stress (3 Pa for The lower limit is set stress is made large enough. An increase in yield by the requirement that the slurry remain stable. stress can be achieved most simply by a concentrat-

surface properties

As an example suppose a coal slurry has a maximum of the colloidal particles can be altered by pH changes or by addition of flocculants to increase particle size of 1.5 mm and an extrapolated yield stress of 2.6 Pa at 48% weight concentration (40% flocculation and so increase the yield stress. by volume). Equation A7 indicates a minimum recause quite dramatic changes in the yield stress. noted by Elliot & Glidden this latter effect can quired yield stress of 1.62 Pa so the slurry is Whether pH or flocculant will affect a slurry of a colloidal slurry is proportional to D.G. Thomas (1963) suggests that the yield depends entirely on surface chemistry of the

slurry and so each slurry needs to be treated

concentration cubed although experiments by individually. the author have often indicated greater dependence up to the seventh power of concentration. In some cases pH changes or the addition of dewould mean that the allowable concentration limits the third power for demonstration purposes this stress so that an economic pumping pressure can be flocculants may be needed to decrease the yield for the slurry under consideration would be between obtained. 34% by volume (43% by weight) and 42% by volume (50% by weight). The main thrust of Elliot & Glidden's paper was 8 THE CONCEPT OF DENSE PHASE TRANSPORT bution to achieve a stable slurry with minimum concerned with adjustment of particle size distri- There have been a number of papers published which pressure gradient. There can be little doubt that suggest that it may be advantageous to pump at adjustment of the size distribution to ensure maximum packing density will produce a less viscous higher concentrations of around 50 to 60% by volume than the current concentrations of between slurry with a lower yield stress. But the stabil- 25 to 45% by volume and indeed the Ohio coal pipethen mean that a lower top size will be required ity theory presented here suggests that this could line did pump coal at 58 to 60% concentration by if stability is to be maintained. weight for some time (Halvorsen (1976)). notable papers are those due to Bantin and Streat and Elliott & Glidden (1970). It is hoped that the concept of dense phase these was concerned with coarse sand (around 0.7 unusual happens when the concentration is increased can now be seen in proper perspective. mm) of narrow size distribution with no particles less than about 0.1 mm. They found that this could above 50% by volume. explained by equations A6 and A7. The stability can still be be pumped at concentrations above 55%. to be aware that transportation further pursued this line. Bantin (1972) and Streat and Televantos (1976) have However, as mentioned les as a stable slurry is achieved at the expense previously, the pressure gradients in this type of of higher pressure gradients. flow are extremely high, some 50 to 100 times applications this rapidly becomes uneconomic once usual practice, and so it would only be of use for mentioned previously. the maximum particle size exceeds For short distance applicatvery short distances. taken by Elliott and Glidden where they found Another line of approach was ions it may well be warranted. by pumping a wide size distribution coal slurry 9. THE DESIGN PROCEDURE containing appreciable colloidal size material at concentrations above 50% coal particles up to 12.5 It is now possible to form a design procedure and mm could be transported under laminar flow and the suggested steps for the design of a long distance slurry was stable and could be, stopped and restartslurry pipeline operating in the turbulent flow This is now easily explained. Equations A6 regime are outlined below: andd A7 in the Appendix indicate that provided the yield stress is made sufficiently high any top 9.1 Determine Maximum Particle Size size particle can be supported and so the slurry But as discussed previously there is The maximum economic velocity limit and the consethe maximum yield stress due to economic quent limit on maximum yield stress of about 4.5 Pa shown that limits of about 3 Pa (T considerations of pumping pressures. ent flow operation or about 7.5 Pa for a completely ) for turbul- It was been i Ted a tory requires do equat on A that that the maxi-

mum particle size be no greater than about 0.3 mm

stable coal slurry operating in laminar flow are for iron ore and about 3mm for coal.

The Institution of Engineers, Australia

Page 9

LONG DISTANCE SLURRY PIPELINES - Thomas 27

(visually through transparent viewing section). turbulent transition was observed at 1.4 ms At stress, T ranging t*about 3 Pa for a coal slurry. of about 4.5 Pa an iron or slurry Both 1.3 ms steady laminar flow without deposition was these depend slightly on the pipe diameter. Note observed. At 0.8 ms a bed of solids formed. Thu s that the turbulent pressure gradient is not greatly 0.8 and 1.3 ms the critical deposit velocity lay somewhere between, well below the laminar/turbulent dependent on differ for different solids S.G. 's. the yield stress although it will As an example, transition velocity. The same slurry, at the same Piercy (1976) has stated that the yield stress for concentration, in the larger pipe behaved differentthe Bougainville copper concentrate slurry (solids ly. turbulent flow conditions. In this case deposition occurred under By scaling up from the S.G. 4.3) is maintained below a maximum limit of 4.4 Pa which is in agreement with the above stated laminar flow results obtained in the smaller pipe limits. The lower limit for stability depends on (see Bowen (1961/1) it was estimated that the the maximum particle size. pipe was about 1.0 ms laminar/turbulent transition velocity in the large, although laminar flow If the slurry is completely stable, operation withconditions could not be realised. in the laminar regime is feasible. For example the

Rugby limestone pipeline in England operates under

The slurry of Figure 3 illustrates typical pseudolaminar conditions. (Shriek et al (1973/1) In

observed by this case the same economic limits on maximum

the author with a number of different slurries. pressure gradient apply so that the upper limit turbulent flow is suddenly stopped the can be raised as shown in Figure 4. This limit

Sustained laminar would be about 2.5 times the previous limits ie.,

is only possible in small pipe sizes and in T = 11 Pa for iron ore down to 7.5 Pa for large pipes deposition occurs under conditions of coal. turbulent flow at velocities above the laminar/ turbulent transition velocity. For pipe sizes Operating point for both large enough for deposition to occur under laminar and turbulent flow conditions of turbulent flow the deposit velocity will vary as D as for a type 3 slurry. priori whether deposition will occur under laminar The author knows of no method of predicting a Upper limit for economic Laminar operation - - or turbulent flow conditions. For rapidly settling pseudo-stable slurries it would therefore seem necessary to perform pipe loop tests in increasingturbulent regime. Once this size pipe is reached 1y larger pipe sizes until deposition occurs in the obtained by assuming it varies as DO.1 the deposit velocity in larger pipe sizes can be as for a ДP/4L Upper limit for economi turbulent operation

type 3 slurry. Log

5.3 The Velocity Should Not Exceed About 1.75 ms l

for Economic Reasons Lower limit for Operation above 1.75 ms is generally uneconomic stability due to the rapid increase in pressure. an upper limit on the yield stress. uncertainty associated with whether or not laminar 0.5 1 2 flow is possible it is generally desirable that the pipeline be operated in the turbulent regime. Velocity (ms"') To ensure that operation is fully within the to allow for possible variat- Figure 4. Yield stress limits for

laminar and turbulent flow

prediction methods the transition velocity should ion in slurry properties and inaccuracies in operation. not exceed about 1ms a safety margin of 0.75 ms Note that this, then allows 6. CONSEQUENT LIMITS ON MAXIMUM PARTICLE SIZE turbulent regime as discussed above. for the possibility of deposition occurring in the For example Considering only turbulently flowing slurries, for the deposition velocity for the slurry of Figure 3 they are by far the most common, the upper limits

approaches the usual operating velocity range of that the maximum possible particle size for a T = 4.5 Pa for iron ore and 3 for coal mean

of 1.5 to 1.75 ms for pipe diameters greater than stable slurry is (from equation A7 of the Appendix) about 1000 mm so for pipes smaller than this there about 0.3 mm for iron ore ranging up to about 3 mm would be no problem if design were based on the laminar transition velocity. In short, as long for coal. Of course a slurry can operate with the slurry does not compact very fast on shutdown maximum particle sizes considerably less than these limits and may have to to prevent deposition (say within a few minutes), the design can be during operation but there is a limit to how fine based on the laminar transition velocity. the particles can be ground for pipelining due to fast compacting pseudo stable Large diameter pipes it would be wise to perform the rapidly increasing yield stress and hence small pipe loop tests to check whether deposition pressure gradient as size is reduced due to increasing colloidal effects. For example the work of does occur in the turbulent regime, and then scale- Friend and Hunter (1971) suggests that the yield bB.by assuming the deposit velocity varies as of the particle size for colloidal stress increases as the square of the reciprocal size particles. Figure 4. The allowable limits on T Note that by plitting D AP/4L versus V Table 1 gives some details of nine existing long distance slurry pipelines together with the maxi- Figure 4 is applicable to all large pipe sizes. The upper limit dictated by the 1 ms maximum limit on mum economic particle size as calculated above. In transition velocity represents a maximum yield every case the maximum particle size is less than

• Chemical Engineering in Australia, 1977

Page 10

28 LONG DISTANCE SLURRY PIPELINES - Thomas

ranging from one fifth to four fifths of the limit. the calculated maximum limit the actual values of Elliott & Glidden's paper will reveal reasonable for long distance applications. A study The economic limits on maximum particle size long distance slurries would appear to lie between for extrapolated yield stress ensuring a stable slurry that the 20% and 100% of the value given by equation A7 using T equal to 4.5 Pa for iron ore ranging with the 12.5 mm top size was around 10 Pa. (at 59% concentration by weight). This is in close

down toes o Pa for coal. agreement with the 1l Pa limit indicated by equation A7. Note, however that it is higher than

LIMITS ON CONCENTRATION omic for long the 7.5 Pa economic limit and so may prove unecondistances. Similarly for a 3.2 mm Most reports on slurry pipeline installations top size they found that a 57.2% by weight concenspecify upper and lower limits centration was not staffe. tration was stable ( T = 4.6 Pa) while 53% con- The yield stress of the (1976) mentions the limits range of concentration. For example Halvorsen latter is difficult to ascertain but they did line as being 50% and 44% by weight. with every 5% increase in concentration by weight. state that the yield stress approximately doubled the reasoning outlined above it is possible to for these limits should now be obvious and following It is remarkable that equation A7 indicates a This would suggest a yield stress of about 2.7 pa. calculate the allowable limits for any particular value of 2.9 required for stability. The upper limit is set by the necessity to stay Thus it can be seen that slurries of virtually any top size can be made stable providing below the maximum economic yield stress (3 Pa for stress is made large enough. coal, 4.5 Pa for iron ore). The lower limit is set stress can be achieved most simply by a concentratby the requirement that the slurry remain stable. Alternatively the surface properties As an example suppose a coal slurry has a maximum of the colloidal particles can be altered by pH changes or by addition of flocculants to increase particle size of 1.5 mm and an extrapolated yield stress of 2.6 Pa at 48% weight concentration (40% flocculation and so increase the yield stress. by volume). Equation A7 indicates a minimum renoted by Elliot & Glidden this latter effect can cause quite dramatic changes in the yield sfress. quired yield stress of 1.62 Pa so the slurry is Whether pH or flocculant will affect a slurry stress of a colloidal slurry is proportional to D.G. Thomas (1963) suggests that the yield depends entirely on the surface chemistry of the volume concentration cubed although experiments by individually. slurry and so each slurry needs to be treated the author have often indicated greater dependence up to the seventh power of concentration. In some cases pH changes or the addition of dewould mean the third power for demonstration purposes this that the allowable concentration limits stress so that an economic pumping pressure can be flocculants may be needed to decrease the yield for the slurry under consideration would be between obtained. 34% by volume (43% by weight) (50% by weight). The main thrust of Elliot & Glidden's paper was 8 THE CONCEPT OF DENSE PHASE TRANSPORT bution to achieve a stable slurry with minimum concerned with adjustment of particle size distri- There have been a number of papers published which pressure gradient. There can be little doubt that suggest that it may be advantageous to pump at adjustment of the size distribution to ensure maximum packing density will produce a less viscous higher concentrations of around 50 to 60% by volume than the current concentrations of between slurry with a lower yield stress. But the stabil- 25 to 45% by volume and indeed the Ohio coal pipethen mean that a lower top size will be required ity theory presented here suggests that this could line did pump coal at 58 to 60% concentration by if stability is to be maintained. weight for some time (Halvorsen (1976)). are those due to Bantin and Streat It is hoped that the concept of dense phase flow (1970) and Elliott & Glidden (1970). seen in proper perspective. mm) of narrow these was concerned with coarse sand (around 0.7 size distribution with no particles unusual happens when the concentration is increased less than about 0.1 mm. They found that this could above 50% by volume. The stability can still be be pumped at concentrations above 55%. explained by equations A6 be aware that transportation Bantin (1972) further pursued this line. and Streat and Televantos (1976) have However, as mentioned les as a stable slurry achieved at the expense previously, the pressure gradients in this type of of higher pressure gradients. applications this rapidly becomes uneconomic once flow are extremely high, some 50 to 100 times the maximum particle size exceeds the limits usual practice, and so it would only be of use for mentioned previously. For short distance applicatvery short distances. Another line of approach was ions it may well be warranted. taken by Elliott and Glidden where they found that by pumping a wide size distribution coal slurry 9. THE DESIGN PROCEDURE containing appreciable colloidal size material at mm could be transported under laminar flow and the concentrations above 50% coal particles up to 12.5 It is now possible to form a design procedure and slurry was stable and could be stopped and restartsuggested steps for the design of a long distance slurry pipeline operating in the turbulent flow This is now easily explained. Equations A6 regime are outlined below: andd A7 in the Appendix indicate that provided the yield stress is made sufficiently high any top 9.1 Determine Maximum Particle Size size particle can be supported and so the slurry made stable. But as discussed previously there is The maximum economic velocity limit and the consea limit to the maximum yield stress due to economic quent limit on maximum yield stress of about 4.5 Pa ent flow operation or about 7.5 Pa for a completely considerations of pumping pressures. It was been shown that limits of about 3 Pa (T ) for turbul- Ted for equores dom equa3 on 40 cat the maximum particle size be no greater than about 0.3 mm stable coal slurry operating in laminar flow are for iron ore and about 3mm for coal.

The Institution of Engineers, Australia

Page 11

LONG DISTANCE SLURRY PIPELINES - Thomas 29

9.2 Perform Settling Tests to Determine fluid having a density equal to the slurry density Stability. and a viscosity equal to the Bingham plastic

viscosity. Cheng (1970) suggests that this method

A sample of the commodity reduced down to a size overpredicts by about 12%. distribution such that the maximum particle size is less than the above limits can now be made up 9.6 Determine Transition and/or Deposit Velocity into a slurry of typical concentration. The usual concentration required to obtain reasonable The intersection of the laminar and turbulent flow solids throughput range from 25% by volume for curves gives the transition velocity for the pipe iron ore to 50% for coal. Slurries of about this size under consideration. This should be less concentration should be mixed and then allowed to than about 1ms If this is so the slurry would settle for a day or so. if they settle into a compacted homogeneous mix- If they do not settle, or be suitable for long distance pipelining at about 1.5 to 1.75 ms For a fast compacting pseudoture, they are suitable as far as stability is stable slurry in large diameter pipes deposition concerned. may occur in the turbulent regime. In such cases

pipe loop tests may be desirable.

9.3 Determine Yield Stress for Start Up Calculation. 9.7 Investigate Possible Changes in Slurry

Properties.

The yield stress of these stable or pseudo-stable

Upon testing the originally selected size

indicated by equations A6 or A7. But they may be distribution and concentration the slurry may be this and indeed might be too found to be quite suitable. However more likely to permit startup after shutdown. it will not be perfect. There are two stable the slurry as is possibilities:

should be tested for yield stress. If the slurry has compacted the top layer of water should be 9.7.1 The slurry may not be stable. If this is drained off and then only the compacted portion the case equations A6 or A7 indicate two alter- In both cases the measured yield stress natives. Either the particle size can be reduced must be low enough to ensure that start up is or the yield stress can be increased. This latter As mentioned previously the relevant option can be achieved by either an increase in yield stress is the extrapolated yield stress and concentration or by a change in the surface this is best obtained using a tube viscometer. chemical properties to increase flocculation. The (e.g. Wazer et al (1963)). However a rotational first alternative would be the most sensible since viscometer can be used if it operates at this means that a higher solids throughput can be sufficiently high shear rates. Alternatively a achieved. The steps 9.3 to 9.6 can then be rotational viscometer such as the Brookfield repeated. instrument can be used to obtain the static yield stress which as is explained in the Appendix can 9.7.2 The slurry is stable but the yield stress then be multiplied by about 5 times to get an is too high. This can be reduced by either approximate value for Tex. decreasing the concentration or by possible

de-flocculation. The latter option is obviously

9.4 Obtain Laminar Flow Curve. the most attractive so that solids throughput can

be maintained but it may not always be chemically

Next the laminar flow curve relating pressure or economically possible. gradient to velocity for the pipe sizes of interest needs to be determined so that scale-up by 10. SPECIFIC ENERGY CONSUMPTION AND SOLIDS method of Bowen (1961/1) can be performed. THROUGHPUT. requires tests on the original mixed slurry which for a pseudo-stable slurry is at a lower concen- For a slurry of any given maximum particle the compacted portion which was testhas been shown how upper and lower limits can be ed above for start up purposes. Some pseudoplaced on the allowable concentration. stable slurries may compact very quickly after practical importance to have knowledge of the agitation ceases which can make it difficult to effect that operation at different concentrations obtain meaningful results from a rotational has on the specific energy consumption and on the viscometer. Such slurries should be tested in solids throughput. tube viscometer where settling is no slurries can be tested in a rotational viscometer. Slowly compacting or fully stable Figure 5 • Knows a plot of specific energy (Joules ) versus solids throughput (tonnes/year) If the flow curve for the full with concentration, volume flow rate (or velocity), rates of interest cannot be obtained and pressure gradient as parameters. This has be a serious limitation since, as indicated by been prepared for the following design situation. Figs 2 and 4, for large pipes the flow curve is dependent almost solely on the extrapolated yield coal S.G. 1.40 If a rotational viscometer is used but Maximum particle size 1.4 mm does not permit sufficiently higher shear rates to Pipe diameter 450 mm obtain the extrapolated yield stress an approximate Tex = 3Pa at C = 45%

static yield stress by about 5. Tex can be found by multiplying the p1 = 0.015 ke m-1 5-1 (15 centipoise)at C = 45%

9.5 Obtain Turbulent Flow Curve esth Tonsideration of figure here of he vary as co in the range of inter-

The next step is to obtain the turbulent flow available region of operation is bordered: curve. This is also best obtained from tube (a) to the left by the need for the concentration viscometer tests or pipe loop tests (scaled up (and hence the yield stress) to be sufficiently using the Bowen (1961/2 method). However if these high to ensure stability. are not available a reasonable estimate can be made by considering the slurry as a Newtonian (b) to the bottom by the need to operate at a

Chemical Engineering in Australia, 1977

Page 12

30 LONG DISTANCE SLURRY PIPELINES - Thomas

175

final design and optimisation it would be prefer-

Pipe Diameter 450 mm able to have the results of pipe loop tests. Also

for pseudo-stable slurries which compact very fast, pipe loop tests may be necessary to determine if

km- 1, - 1 kg 150- =•35 (V=_2.2) — diameter). deposition occurs in the turbulent regime. is being considered. (say in excess of 300 mm especially desirable if a very large diameter pipe This is

(Joules The author thanks M.D. Research Company Pty. Limited for permission to publish this paper. 12 ACKNOWLEDGEMENTS

Energy Consumption 100 9=.25 (V=1.57 2=30_ (V=1.89). 9.28 (V =1.75) velocity Journal Vol. 13, PP 1193-1196. Spherical Particles Engineering, October, p. 68. ANSLEY, R.W. and SMITH, J.N. (1967) 13 REFERENCES Anonymous (1975) New Slurry Pipeline, Mechanical in a Bingham Plastic. A.I.Ch.E. Motion of

sport 1 Conference, Coventry, England, paper GI

Specific 75 50 Predicted laminar transition Ur- 6 of Solids - Water Mixtures in Pipelines. Hydrotran- BANTIN, R.A. and STREAT, M. (1970) Dense-Phase Flow Systems. BOWEN, R.L. (1961/2) Designing Turbulent Flow BOARDMAN, G. and WHITMORE, R.L. (1961) The static Chemical Engineering, June 12, PP BOWEN, R.L. (1961/1) Designing Laminar-Flow Systems British Hydromechanics Research Association. Chemical Engineering, July 24, pp 143-150 243-248 Solids throughput (millions tonnes/year) Measurement of Yield Stress, Laboratory Practice,

November, PP 782-785.

Figure 5. Performance graph for a typical coal CHENG, D.C.H. (1975) Pipeline Design for Nonslurry. Newtonian Fluids, The Chemical Engineer, Parts 1&

Q = volume flow rate (m?. P = pressure gradient (RPa. ken"!, CHENG, D.CH (1970) A Design Procedure for Pipeline 2, Sept & Oct.

V = velocity (m. s-1) Flow of Non-Newtonian Dispersed Systems, Hydro- C = concentration (volume %) transport 1 Conference, Coventry, England, paper

J5 British Hydromechanics Research Association.

velocity sufficiently above the predicted laminar transition velocity to ensure operation in the COURATIN, P. (1969) Tailing Disposal, World Mining, turbulent regime. A safety margin of 0.3 ms-1 has May, PP 38-43. been assumed and is represented by the hatched area. DINA, M.L. (1976) Operating Experiences at the 1580 (c) to the top by limits on either the maximum Mw Coal Slurry Fired Mohave Generating Station, economic velocity or the maximum pressure available. First Int. Conference rd slurry Transport Associat- Columbus, Ohio, Feb. Slurry Transportation, This plot shows that both for a given solids ion, Washington, D.C., U.S.A. throughput and a given maximum pressure it is more DURAND, R. (1953) Basic Relationships of the Transefficient to operate at the highest possible portation of Solids in Pipes - Experimental Researconcentration and the lowest possible velocity within the limitations imposed by the above considerch. Proceedings 5th Minneapolis Minnesota Inter-

national Hydraulic Convention, PP 89-103 International Association for Hydraulics Research.

11 CONCLUSIONS ELLIOT, D.E. and GLIDDEN, B.J. (1970) Hydraulic It has been shown how the three design considerattransport 1 Conference, Transport of Coal at High Concentration. _Coventry, England, paper Hydroions of stability on shutdown, avoidance of deposi- G2 British Hydromechanics Research Association. tion during operation, and economic limit to maximum velocity and pressure gradient can dictate the FRIEND, J.P. and HUNTER, R.J. (1971) Plastic Flow Behaviour of Coagulated Suspensions Treated as a suitable for long distance slurry transportation. available range of particle sizes and concentration Repeptisation Phenomenon. J. Colloid and Interface To aid in the consideration of slurry stability a Science Vol. 37, No. 3 pp 548-556. simple theory has been developed based on the con- GAY, E.C., NELSON, P.A. and ARMSTRONG, W.P. (1969) cept of the largest particles being Flow Properties of Suspensions with High Solids the yield stress of the slurry. Concentration, A. I.Ch.E. Journal, Vol. 15, n 6, design considerations and stability November, pp 815-822. which are in excellent agreement with all informatin limits on particle size and slurry GOVIER, G.W. and AZIZ, K. (1972) The Flow of ion available to the author concerning existing New York. Complex Mixtures in Pipes, Van Nostrand Reinhold, long distance slurry pipelines.

HALVORSEN, W.J. (1976) Slurry Pipeline Hydraulics

Using the above concepts a series of design steps Improved. The Oil and Gas Journal, March 22, 1976, are outlined which should serve as a design propp 62-66. cedure for long distance pipelines. This can be McDERMOTT, W.F., DAVIS, R.A., COWPER, N.T. and based entirely on bench scale tests although for

The Institution of Engineers, Australia

Page 13

LONG DISTANCE SLURRY PIPELINES - Thomas 31

WASP, E.J. (1969) The World's First Long Distance port 2 Conference, Coventry, England. Paper HI, Iron Ore Slurry Pipeline. Mining Engineering, British Hydromechanics Research Association. January, PP 86-89. VOCADLO, J.J. and CHARLES, M.E. (1972) Prediction MCNAMARA, E.J. (1976) Operational Problems with a of Pressure Gradient for the Horizontal Turbulent 69 mile Copper Concentrate Slurry Pipeline. Flow of Slurries. Hydrotransport 2 Conference, Hydrotransport 4 conference, Banff, Canada, Paper Coventry, England. Paper Cl, British Hydro- F3, British Hydromechanics Research Association. mechanics Research Association. MICHAELS, A.S. and BOLGER, J.C. (1962/1) Settling Rates and Sediment Volumes of Flocculated Kaolin Fundamentals, Vol. 1, No. 1, Feb., Pp 24-33. Suspensions, Industrial and Engineering Chemistry R.B. NASP, E.J. 1970) Deposition Verties, AUDE, Transition and JACQUES, MICHAELS, A.S. and Bolger, J.C. (1962/2) The Velocities, and Spatial Distribution of Solids in Plastic Flow Behaviour of Flocculated Kaolin Slurry Pipelines, Hydrotransport 1 Conference, Suspensions, Industrial and Engineering Chemistry Coventry, England, Paper H4, British Hydromechanics Fundamentals, Vol. 1, n 3, August, PP 153-162. Research Association.

NOYAN, K., AKSOY, S. JENSEN, J.H. and WRIGHT, P.B. AUDE, T.C., SEITER, R.H. and THOMPSON,

(1974) Concentrates. Hydrotransport 3 Conference, The KBI Parallel Pipelines for Sulphide in the Turbulent Regime. Int. Symp. on Solid- Colorado, U.S.A., Paper B2, British Hydromechanics Liquid Flow in Pipes, Uni. of Penn., Proceedings Research Association. available in book "Advances in Solid-Liquid Flow PIERCY, P. (1976) Hydraulic Transport of Copper in Pipe and its Application Pergamon, Oxford (1971)., edited by I. Zandi, Concentrate At Bougainville. Transport of Minerals Newcastle Chemical Engineering Group, University in the Process Industries, Half Day Symposium, MAZ (1963) Viscosity and Plow Measurement - A J.R., LYONS, J.W., KIM, K.Y. and COLWELL, of Newcastle, July, pp 29-35. Laboratory Handbook of Rheology, Wiley, New York. RAUDKIVI, A.J. (1975) Private Communication. Uni. APPENDIX of Auckland, N.Z. SCHRIEK, W., SMITH, L.G., HAAS, D.B. and HUSBAND, Criteria for Stable Slurry W.H.W. (1973/1) Experimental Studies on Solids pipelining of Canadian Commodities Report II. It has been shown that to permit restarting of a Experimental studies on Hydraulic Transport of stopped pipeline the slurry must be either complete Limestone. Saskatchewan Research Council, Canada. ly stable even when stopped or if it does settle

it must compact with no preferential settling of

W.H.W. (1973/2) Experimental Studies on Solids SHRIEK, W. SMITH, L.G., HAAS, D. and HUSBAND, the coarser particles. termed pseudo-stable. What governs whether a This latter case has been Pipelining of Canadian Commodities Report II! Experimental Studies on the Hydraulic Transport of slurry will fit these requirements? Iron Ore. Saskatchewan Research Council, Canada. Consider the largest particles in a slurry Pipelining of Canadian Conmodities Report V SHRIER, (1973/3) Experimental Studies on Solids W.H.W. SMITH, L.G., HAAS, D. and HUSBAND, immediately after shutdown. resistance force due to the yield stress of the fall if the nett gravitational force exceeds the These particles will Experimental Studies on the Hydraulic Transport of slurry. This then is the criterion for a slurry Coal - Saskatchewan Research Council, Canada. to be stable (either completely or pseudo) - the SHOOK, C.A. (1974) Experimental Studies on Solids yield stress of the slurry must exceed the yield Pipelining of Canadian Commodities Report I stress required to support the coarsest particles. Saskatchewan Research Council, Canada. If the stress everywhere on the surface of a STREAT, M. and BANTIN, R.A. (1972) Mechanism of sphere of diameter d is T the resistance force in Hydraulic Conveying at High Concentrations in the vertical direction can be shown to be given by Vertical and Horizontal Pipes. Hydrotransport 2 (A1) Conference, Coventry, England, paper B2, British Hydromechanics Research Association. The exact stress distribution on a sphere immersed STREAT, M and TELEVANTOS, Y. (1976) Pilot Plant in a Bingham plastic is not known so that equation at High Concentration in Pipelines. Hydrotransport Studies of Hydraulic Conveying of Coarse Materials (Al) needs to be replaced by. F = к*п? 2т (A2) 4 Conference, Banff, Canada, Paper F2 Hydromechanics Research Association. British To prevent settling of the particle this force must

exceed the nett gravitational force acting on the

THOMAS, A.D. (1976) Scale-Up Methods for Pipeline particle. For a solid particle immersed in a Transport of Slurries. International J. Mineral fluid the nett gravitational force is determined Processing, Vol. 3 pp 51-69. by allowing for the buoyancy effect of the fluid THOMAS, A.D. (1975) Unpublished data at M.D. as per Archimedes' principle. However in a stable Research Company. slurry at rest the coarse particles are supported by the plastic yield stress and Archimedes' THOMAS, D.G. (1961) Transport Characteristics of principle may not be valid for such slurries if Suspensions: II Minimum Transport Velocity for the pressure cannot be transmitted hydrostatically. Flocculated Suspensions in Horizontal Pipes. Because of this Ansley & Smith (1967) state that A. I.Ch.E. Journal Vol. 7, No. 3, PP 423-430. buoyancy should not be allowed for. However THOMAS, D.G. (1963) Transport Characteristics of Boardman and Whitmore (1961) found that for a flocculated China clay suspension Archimedes' Suspensions VII. Relation of Hindered-Settling principle held to within about 2.5% which was Floc Characteristics to Rheological Parameters, insignificant variation compared with the experi- A. I.Ch.E. Journal, Vol. 9, n 3, May, PP 310-316. mental error involved. E.J. (1972) Slurry Pumps - a survey. Hydrotrans- THOMPSON, T.L., FREY, R.J., COWPER, N.T. and WASP, The apparent applicability of Archimedes' principle

Chemical Engineering in Australia, 1977

Page 14

32 LONG DISTANCE SLURRY PIPELINES - Thomas

with flocculated suspensions is perhaps not really slurry. In fact it appears to depend on the size collodal surprising. In a flocculated suspension the distribution of the slurry. A slurry having all form "puffy" flocs. At high concentrations and size particles are attracted together to colloidal size material such as clay gives Tex/Tst conditions of low or zero shear rate these flocs values approaching 1 while a slurry such as the It is this structure which supports the coarse cluster together to form an aggregrate structure. particles in a stable slurry. above coal slurry having only about 15% colloidal values of around 5. size (less than.01 mm) particles gives T Physically s is ean be explain conceivable that each large particle might only be ed by assuming the Brookfield instrument measures supported at two or three points with a fluid the yieldstress due to the colloidal size particllayer occupying most of the immediate area es only whereas the extrapolated yield stress obsurrounding the particles. Under such conditions tained using the tube viscometer depends both on principle would be expected to hold, the colloidal effects as well as a purely mechanicto some degree at any rate. Based on the evidence al component due to any coarse particles present of Boardman & Whitmore the nett gravitational similar to that investigated by Gay et al (1969). force has been calculated allowing for the buoy- In the Brookfield instrument with the bob rotating ancy effect. (It will be shown later that if in a container of effectively infinite diameter the buoyancy had not been allowed for the stability slurry is not constrained and so, possibly helped particle sizes of commercial slurries). To prevent theory would be in poor accord with the reported by centrifugal migration of the coarse particles, the bob only "sees" the colloidal material.

the tube viscometer the slurry is constrained and so the coarse particles contribute to the extra-

= > ≥ na (Pp -Pm)8 (A3) polated yield stress.

where Pp is the density of the particle Smith (1967). Equation A5 was obtained from the data of Ansley & They employed tomato sauce (catsup) and Pm is the density of the mixture whose properties were obtained by fitting Bingham's Equating A2 and A3 equation to tube viscometer data, i.e., their T > - 2d Pp- Pm8 3 кП (A4) solid particles present so that from the above sauce would be expected to have yield stress is equivalent to Telew* if any, coarse Ansley and Smith (1967) considered spheres falling field instrume st had been measured using a Brookthe result would be expected to be in Bingham plastic slurries. They give pertinent approximately equal to Tex• Thus equation A5 can data for spheres which fell slowly and another

which was supported in a particular slurry. this K can be found equal to 2.3 so that equation From Tst >0.092 a (Pp - P)g (A6 )

(A4) becomes For a slurry consisting almost entirely of colloid-

T > 0.092 d (Pp - Pm8 (A5) al size material with perhaps only a few percent coarse particles, Tst in equation A6 could be re-

This then is the stability criterion. d will be placed by Tex However in the case of the more the diameter of the largest particle present. usual commercial slurries having a wide size

available equation A6 can be distribution T = 5 T so that if Tex only is

Before applying this equation to pipeline design replaced by

of slurries. it is necessary to consider a rheological aspect Figure Al shows an experimentally Tex > 0.46 a (Pp- Pm8 (A7)

determined plot of shear stress versus apparent shear rate obtained using a tube viscometer for a This is only an approxiamtion but in the author's distribution almost identical to that reported by coal slurry tested by the author and having a size Tst for the slurries of commercial type size ex'S generally between 4 to 6 times Dina (1976) for the Black Mesa coal slurry. This can be converted to a plot of shear stress versus tained from a true shear rate plot. distribution. It should be noted that Tex is ob- If the straitrue shear rate and this is shown in Figure A2 ght line portion of the apparent shear rate (8V/D) along with similar results obtained using a Brookplot is extrapolated to zero shear the intercept field rotational viscometer with a cylindrical bob will be 4 Tex/3. (dashed line on Figure A1). rotating in a very much larger container. straight line portion of this plot is extrapolated Equations (A6) and (A7) therefore provide the reto zero shear rate the intercept gives the extraquired stability criterion.Equation Abis the prewith the plastic viscosity, nol = polated yield stress, T ferable one to use if Tst is available. As examples, A typical top size for a coal slurry provide the Bingham plastic parameters. (coal S.G. 1.4) might be 10 mesh (2.0 mm). For a Bingham model is seen to fit this slurry for shear ates above about 100 sec The full line in indicate a minimum required yield stress of Tst = 40% by volume concentration equations A6 and A7 Figure Al is the shear stress versus apparent 0.43 Pa or Tex = 2.1 Pa. Piercy (1976) has proshear rate relationship calculated according to vided information relating to the Bougainville the Bingham model. At very low shear rates the copper concentrate pipeline. This slurry, solids measured shear stress is seen to be less than the S.G. 4.3, top size 0.208 mm is pumped at between Bingham parameter, Tex, but for pipe sizes and 55 and 70% concentration by weight. Taking the velocities of commerical interest this is immaterlower of these values equations A6 and A7 indicate ial. Note that the yield stress obtained with the a required minimum yield stress of Tst = 0.48 Pa Brookfield instrument, termed the static yield or Tex = 2.4 Pa. This latter value is remarkably stress, Tst, is only 2.6 Pa ie., about one fifth close to the minimum permissible yield stress of of Tex,. However, similar tests on various pure 2.1 Pa as stated by Piercy. Brookfield instrument almost equal to the Tex clay suspensions have given Tst values from the It was noted earlier that there is some doubt as to values obtained in the tube viscometer. Thus the the validity of the buoyancy correction with ratio Tex/ Tst depends on the properties of the plastic fluids. imental evidence of Boardman & Whitmore (1961) it However on the basis of the exper-

The Institution of Engineers, Australia

Page 15

LONG DISTANCE SLURRY PIPELINES - Thomas 33

was allowed for and the stability criteria expressbuoyancy correction are much closer to the actual values providing strong evidence that the stability ed in equation A6 and A7 were consequently developcriteria developed here, allowing for the buoyancy ed. The limits on maximum particle size, calculatcorrection, is most realistic. ed using these equations and the limitations on yield stress discussed in Section 5, can now be It should be noted that the stability criteria different commercial slurries. compared with the actual maximum particle size of Referring to Table developed here cannot distinguish between a fully stable slurry and pseudo-stable one. To be 1 the ratio of calculated maximum particle and magnetite is 10:1 comsizes completely stable and hence capable of being

of, for example, coal pared to the actual ratio of the Black Mesa coal operated under laminar flow conditions a slurry

to the Savage River magnetite of 23:1 or using Ohio coal 14:1. If the buoyancy correction had not the given by equations A6 may have to have a yield stress well above that or Al, especially for

been made these ratio would have been 2.4 and 1.4 operation in large diameter pipes as was discussed in Section 5.2.

respectively. Obviously the values using the

60

60

50

50

40 40

30 (Pa) 30

20 10 20 10 78000-00.: Brookfield tube viscometer

500 1000 1500

500 1000 1500

Shear Rate (sec 1,

Apparent Shear Rate 8V/D (sec 1) Figure A2. Coal slurry test results from

Figure Al. Coal slurry test results from tube viscometer and rotational tube viscometer. Shear stress viscometer. Shear stress versus versus apparent shear rate. true shear rate.

Chemical Engineering in Australia, 1977